T H E JOURNAL OF I N D U S T R I A L A N D E N G I N E E R I N G C H E M I S T R Y
May, 1913
389
TABLEI Nitrogen Phos. (PaOs) Moisture Oils DESCRIPTION Per cent Per cent Per cent Per cent From Eubanks Tankard Co. Dry scrap (from 6 sacks) 8.93 6.17 6.48 5.91 From Taft Fish Co. Dry scrap (sample from 525 tons) 8.96 '7.75 6.18 6.81 From Carter's Creek Fish Guano Co. Dry scrap, dried in hot air ard steam driers (from one sack) Fall product 7.70 5.22 11.68 6.62 Cape Charles, Va. From Atlantic Fish & Oil Co. Dry scrap, ground (from 3 sacks) 9.29 6 . 12 7.86 5.38 Cape Charles, Va. From Atlantic Fish & Oil Co. Dust from grinders 8.80 5.21 7.17 7.55 Beaufort. N.C. From Beaufort Fish-scrap & Oil Co. Dry scrap, hydraulic presses, sample from heap 8 . 2 2 S .95 6.13 8.57 Morehead City. N. C. R. W. Taylor. Dry scrap from open heap 8.49 5.95 9.12 8.23 Morehead City, N. C. From Chas. S. Wallace. Scrap, dry, from hydraulic presses 7.76 9.65 8.15 7.56 Lenoxville, N. C. From C. P. Dey. Ground scrap, sulz dried, hydraulic presses. Sample from heap 7 . 8 1 S .85 7.46 7.89 Lenoxville, N. C. From C. P. Dey. Scrap, dry, ground, hydraulic presses. Sample from heap 8.29 9.00 7.00 5.40
No.
LOCATIOK 1 Kilmarnock, Va. 2 Taft. Va. 3 Irvington. Va. 4 5 6 7
8 9 10
AVERAGES ON MOISTURE-FREE FISH SCRAP
11 Crisfield, Md.
From L. E. P. Dennis & Son.
Ground crab shells, used as filler
9.13
i.25
0.00
7.09
3.82
4.55
6.95
2.11
BUREAUO F SOILS
bromate method (Koppeschaar's solution) are equally rapid, accurate, and satisfactory, under proper condiWASHINGTOK, D. C. tions of acidity and phenol concentration, in forming THE EFFECT OF TEMPERATURE, ACID CONCENTRATION the white, flocculent, insoluble tribromphenol. The hypobromite solution is shown t o have the disadvantage AND TIME ON THE BROMINATION OF PHENOL FOR QUANTITATIVE DETERMINATIONS of being unstable when not carefully sealed or kept B y L. V. REDMAN, A. J. WEITH AND I?. P. BROCK in air-tight bottles, and as a consequence requires Received February 26, 1913 restandardization every day. The diluting of the The rapid introduction into commerce of synthetic phenol solution has entirely prevented the influence plastic resistives made from phenol and compounds of the yellow tribromphenolbromide and the formation containing mobile methylenes has made the rapid of the red tetrabromphenoquinone, which have been and accurate determination of phenols of great im- noted as sources of error by earlier investigators.' portance. This paper deals with the simplest of the Errors from these sources were consequently avoided. series, C,H,OH, and will be followed later by methods There remains, however, t o perfect a method for for the determination of the higher homologues. determining phenol, a n investigation into : The fact t h a t seventy-two investigators have con( a ) The effect of acid concentration during the tributed research papers on the determinationof phenols bromination period. is indicative of the trouble experienced in making ( b ) The length of time required for the liberation accurate assays of phenol. Previous investigators of the iodine b y the bromine. have worked upon the theory t h a t the bromination (6) The necessary excess of free bromine t o be used. of phenol is a slow rate reaction and requires time for ( d ) The effects of temperature. completion. To this end they have employed small ( e ) The excess of potassium bromide necessary in volumes of rather concentrated solutions, with con- the bromide-bromate solution. siderable excess of bromine during the bromination The only equipment required for the investigation period. The result is a precipitate of tribromphenol, was a mechanical shaker,%several half-liter grounddense, and almost granular in structure, having often stoppered bottles and standardized burettes. the yellow color of the tribromphenolbromide or The solutions used were a s follows: N / I O sodium containing the separate red specks of tetrabrompheno- thiosulfate, N / I O bromide bromate, N / I O phenol, quinone. As these products1 are not completely 2 0 per cent potassium iodide, hydrochloric acid reduced b y hydriodic acid, the determinations often (sp. gr. 1.2) and a starch solution made by stirring varied b y one or more per cent. 5 grams of starch into a liter of water, heating slowly A complete bibliography of the earlier work is t o until a clear solution is obtained, and allowing the be found in Lloyd's paper' and references to the more gelatinous material t o settle out. recent investigations are found in the researches of The thiosulfate solution was prepared by dissolving Wilkie." I 2 5 grams of sodium thiosulfate (Na,S,0,.5H,O) in Recently, Rhodes and Redmans have shown t h a t if 5 liters of water; 2.76 grams:of:potassium bromate and the concentration of the phenol be approximately 4 3 grams of potassium bromide per liter of solution ' N / I O O during the bromination period, the precipitate constituted the bromide-bromate solution ; a later is a light flocculent, white mass through which the solution in which the potassium bromate was 2 . 7 6 solution can diffuse easily and a s a consequence on grams per liter but the bromide was reduced to 15 thorough shaking the reaction is completed in one grams per liter was found t o be quite as satisfactory minute's time; there is no rapid return of the blue as the 4 3 grams'?per liter. The higher bromide concolor after titration, such as Koppeschaar mentions tent was introduced in the earlier methods to prevent as incident a t times in his method; and from their the formation of tribromphenolbromide,3 but in the numbers a s published the results are accurate t o two dilute solutions is quite unnecessary, as is shown later or three parts in a thousand. These authors have in this paper. The phenol solution was made from the shown t h a t Lloyd's hypobromite and the bromide1 Beckurts, Arch. Pharm., 6, 24, 561 (1886): J . SOC.Chen. I d . , 6,
u. s. DEPARTMENT O F AGRICULTURE
1
Jour. A m . Chem. SOC., 47, 16 (1905).
* J . SOC.Chen. I d . , 30, 398
(1911); 31, 208 (1912). THISJOURNAL, 4, 655 (1912).
546 (1886). 2 THISJOURNAL, 4, 656 (1912). a Lloyd, Jour. A m . Chem. S O C .27, , 15 (190.5).
T H E J O U R N A L OF I N D U S T R I A L A N D E N G I N E E R I N G C H E M I S T R Y
3 90
third distillate at 181-2’ C. of Merck’s c. P. phenol, 1.56 grams being dissolved and made up t o one liter. The 2 0 per cent KI solution was made by dissolving zoo grams of KI in 800 cc. of distilled water. The solution was not made up fresh each day as is frequently recommended. A liter of the solution required only a few drops of N / I O thiosulfate t o reduce the free iodine which was formed in the solution during three weeks. T H E E F F E C T O F ACID C O N C E N T R A T I O N
In each of the following tables the order from left t o right is the order in which the solutions were added for the determinations. The tables in each case show the amount of water added for dilution, the amount of acid added and the resulting normality reckoned for the total volume after the addition of the bromide-bromate solution, i. e . , the acid concentration during the bromination period. The excess bromine calculated and observed are given, and are expressed in percentages of the amount of bromine required t o form tribromphenol of all the phenol present in the solution. The length of time given for the bromination, or the “reaction period,” is expressed in minutes, as is also the time given to liberate the iodine from the potassium iodide b y the excess bromine. The amounts of 2 0 per cent potassium iodide solution added and the amounts of thiosulfate used for titrating back the free iodine are expressed in cubic centimeters and the final column gives thepercentage of phenol found in each experiment. TABLEI-ACID CONCENTRATION Reaction period for B r , 1 min. 3 min. Time to liberate I,
W a t e r , 50 cc. HCI Expt. No. 1 2 3 * 4 5 6 7 8 9 10
11 12 13 14 I5 16 17
C__
Phenol
NorCc. mality 1 0.16 1 0.16 1 0.16 1 0.16 2 0.32 2 0.32 2 0.32 3 0.48 3 0.48 4 0.64 4 0.64 5 0.64 5 0.64 6 0.96 6 0.96 7 1.12 7 1.12
sol. Cc. 15.00 15.00 15.00 15.00 14.97 14.99 14.98 15.00 15.00 15.00 15.00 24.90 24.95 15.00 15.00 15.00 15.00
Br sol. Cc. 15.00 15.00 15.00 15.00 15.28 15.27 15.27 15.00 15.00 15.00 15.00 25.00 25.00 15.00 15.00 15.00 15.00
P e r cent excess of B r Cal.
Obs.
Cc.
Cc.
Per cent of phenol found
1.6 1.6 1.6 1.6 3.6 3.5 3.5 1.6 1.6 1.6 1.6 1.9 1.7 1.6 1.6 1.6 1.6
69.0 60.5 14.0 4.4 6.5 5.7 4.8 1.4 1.5 1.7 1.7 1.9 1.8 1.5 1.6 1.6 1.7
1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
10.35 9.07 2.07 0.65 1.20 0.87 0.75 0.21 0.22 0.25 0.26 0.49 0.44 0.23 0.23 0.24 0.26
31.47 40.77 87.58 97.21 97.09 97.70 98.37 100.20 100.10 99.91 99.88 100.00 99.97 100.08 100.03 100.00 99.88
c _ _
KI sol. Na&03
KI.. . . . . . . . . . . . 20 per cent sol. HCl., . . . . . . . . . . 1.2 sp. gr. Phenol sol..
.....
0.09843 N
Temp..
. . . . . . . . . . . . . . . .22’
C.
The results in Table I show very clearly t h a t the solution must have a n acid concentration of a t least 0.48 N, if the bromination is t o be complete in one minute’s time. An increase in acidity from 0.48 N t o 1.12 N does not, in anyway,affecttheresults. Later, Table V will show that the weaker acid solutions are quite satisfactory if speed is not required, in other words if a longer reaction period than one minute be allowed, complete bromination may be effected in acid solution less than 0.48 N.
Vol. 5 , No. 5
As a n increase of the acid1 t o I O per cent does not affect the accuracy of the results by freeing a measurable quantity of iodine from the hydriodic acid during the determination. 0.84 N acid is recommended as a safety factor Ior securing rapid and accurate results. T H E E F F E C T O F POTASSIUM I O D I D E AND T I M E
Table I1 deals with the length of time and the amount of potassium iodide required to reduce the excess free bromine in the solution with the simultaneous freeing of a proportional amount of iodine. In Experiments 18, 19, 2 0 , 21, the amount of 2 0 per cent. KI solution added was only 0.1 cc. or about I O O per cent more than is required to reduce the excess bromine present. The time given for the potassium iodide to reduce the excess bromine and liberate the iodine was one minute in Experiments 18 and 19,and 3 minutes in Experiments 2 0 and z I . TABLE11-POTASSIUMIODIDE Water 50 cc.
P e r cent excess of B r Expt. No. 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32
HCI Cc. 5 5 5 5 5 5 5 5 5 5 5 5 6 6 6
Phenol sol. Cc. 14.99 14.99 14.99 14.97 14.97 14.98 14.98 14.99 14.99 14.99 14.99 14.99 24.90 24.95 20.00
AND
TIXE
Reaction period for bromine, 1 minute
B r sol. __h_ Cc. Calc. Obs. 15.28 15.28 15.28 15.28 15.28 15.28 15.91 15.91 15.28 15.28 15.28 15.28 25.00 25.00 20.04
3.5 3.5 3.5 3.7 3.7 3.6 7.9 7.9 3.5 3.5 3.5 3.5 2.0 1.8 1.8
3.4 3.4 3.7 3.8 3.9 3.8 7.7 7.7 3.4 3.5 3.4 3.4 2.0 1.8 1.8
Per K I Time to cent of sol. liberate NaZS203 phenol Cc. iodine Cc. found 0.1 1 0.49 100.1 0.1 1 0.50 100.1 0.1 3 0.55 99.82 99.95 0.1 3 0.55 0.5 1 0.57 99.80 0.5 1 0.56 99.81 0.5 3 1.12 100.20 0.5 3 1.14 100.12 1.0 1 0.53 100.05 1.0 1 0.52 100.00 1.0 2 0.53 100.05 1.0 2 0.53 100.05 1.0 3 0.49 100.00 99.97 1.0 3 0.44 5.0 3 0.35 100.02
KI... . . . . . . . . . . 20 per cent sol. HC1.. . . . . . . . . . . 1.2 sp. gr. Phenol sol.. . . . . . 0,09843 N
I n every case the determination is within 0.2 per cent. which is within the limit of error for the reading of the three burettes. Excess of potassium iodide above 0.1 cc. had no effect upon the determination of the free bromine. Three minutes’ shaking to free the iodine gave results which did not vary from the one-minute determinations. This is not in agreement with Rhodes and Redman’s statement that a n error of 0.5 per cent. may be introduced with only a minute’s shaking after adding the potassium iodide. As no account was taken of temperature in their experiments (laboratory temperatures), it is probable that the 1 / 2 per cent error which they record as due t o incompleteness in the freeing of the iodine in one minute may be more properly attributed to incomplete bromination due to low temperatures in the laboratory.
Thiosulfate.. . . . . . . . . . . 0.1 N B r .................... 0.1N Temp.. . . . . . . . . . Approx. 22O C.
T H E E F F E C T O F EXCESS BROMINE
Table 111 records the effect of excess bromine during the bromination period. The experiments show t h a t the phenol present is changed quantitatively into tribromphenol in one minute’s time, without any large excess of bromine, if the proper dilution, acidity and temperatures be observed, e. g., in Experiments 1
Lloyd, Jour. Am. Chem. SOC.. 27, 24 (1905).
33 and 3 4 , 2 per cent excess bromine mas used and the reaction was quite complete. Larger amounts of excess bromine ( 2 . e . , up t o 30 per cent) had no effect upon the determinations. TABLE 111-EXCESS BROXISE \Vater, 50 cc.
Reaction period for bromine, 1 minute Time t o liberate iodine, 1 minute
of "21 e x p t . Cc. 33 34 35 36 37 38 39 40
CO.MBINED E F F E C T O F ACID A S D T I M E
P e r cent. excess of B r
h-0.
6 6 5 5 5 5 5 5
Phenol. sol. Cc.
Br sol. Cc.
24.95 20.00 14.99 14.99 14.98 14.99 14.98 14.99
25 .OO 20.04 15.28 15.28 15.91 15.91 18.29 19.10
Per cent of -*-KI SazC2O3 phenol Cal. Obs. sol. Cc. Cc. found 1.8 1.8 3.5 3.5 7.9 7.9 12.4 29 5
1.8 1.8 3.4 3.4 7,i 7.7 12.1 29.4
K I . , . . . . . . . . . . 20 per c e n t sol. HCI.. . . . . . . . . . . 1.2 sp. gr. P h e n o l . . . . . . . . . . 0.09843 JV
0.1 0.1 0.1 0.1 0.5 0.5 0.5 0.5
0.44 0.35 0.49 0.50 1.12 1.14 1.78 4.36
99.97 100.02 100.10 100.10 100.20 100.12 100.10 100.12
B r . . . . . . . . . . . . . . . . . . . 0.1.V T e m p . . . . . . . . . . . . . . . . . . 22' C.
I t may be noted, however, in determining unknown solutions t h a t the bromide-bromate solution must be added until a slight yellow color, indicating free bromine, remains permanently after shaking. This generally necessitates j to 7 per cent excess bromine, if the yellow color is to be seen easily. To one accustomed .to the method, 2 per cent. excess bromine can be detected. The smaller the amount of bromine in excess the less --ill be the possible loss froni evaporation and such great care is not necessary in preventing the escape of free bromine. T H E E F F E C T O F TEMPER.4TURE
The effect produced upon the determinations b y changes of temperature when the bromination period is one minute is shown in Table ITT,. TABLE IV-CHANGE OF TEZIPERATCRE W a t e r , 50 cc. 5 cc. "21,
of Phenol e x p t . sol. Cc. 41 42 43 44 45 46 47 49 50 51 52
53 54 55 56 57 58
Reaction period f o r B r , 1 min.
. Time to liberate iodine, 1 min. P e r cent excess Br. Sol.
so.
14.99 15.00 15.00 14.97 14.98 14.99 14.99 14.98 14.99 14.99 14.98 14.98 14.99 14.99 14.98 14.98 14.98
Br sol. Cc. 15.25 18.46 18.46 15.28 15.28 15.28 15.28 15.28 15.28 15.28 15.26 15.26 15.28 15.28 15.28 16.55 15.28
7 -
Calc.
Obs.
3.2 25.0 25.0 3.5 3.5 3.5 3.5 3.6 3.5 3.5 3.5 3.5 3.4 3.5 3.5 12.3 3.5
12.5 37.4 27.3 16.7 3.9 4.0 3.9 4.0 3.4 3.4 3.3 3.3 3.3 3.1 3.0 11.0 1.2
K I . . . . . . . . . . . . . 20 p e r cent sol. HCI.. . . . . . . . . . . 1.2 sp. gr. P h e n o l . . . . . . . . . . 0 , 0 9 8 4 3 AT
Temp. at KI sol. Sa29203 e n d of Cc. Cc. titration 1.0 5.0 5.0 1.0 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 1.0 0.1
1.84 5.52 4.03 2.41 0.60 0.60 0.57 0.60 0.49 0.50 0.49 0.50 0.49 0.46 0.44 1.61 0,li
2'C. 2OC. 2'C. 2'C. 17OC. 17OC. 18' C. 18' C. 22°C. 22OC. 26'C,
Zinc. 30° C. 33'C. 34'C. 39'C. 54OC.
shaking. For temperatures ranging from 2 2 ' C. to 30 O C . , the results are very satisfactory and accurate. Above 30° C. the amount of phenol determined is too high. This error is due to increased activity of the iodine a t the higher temperatures in oxidizing the thiosulfate and tetrathionate to sulfate.1
Per centof phenol found 90.88 87.66 97.i4 87.34 99.56 99.50 99.63 99.56 100.10 100.09 100.20 100.10 100.10 100.40 100.62 101.30 102.45
Thiosulfate. . . . . . . . . . . . 0 . 1 N B r. . . . . . . . . . . . . . . . . . . . 0.1 N
At 2 ' C. the bromination is not complete in one minute and no precipitate of tribromphenol is formed. (Later experiments in Table VI will show t h a t a t z o C. complete bromination is effected only after fifteen minutes' shaking.) At 1 7 O and 1 8 the ~ results are still one-half per cent low with the one minute's
I n Table I , i t 17-as shown that the concentration of the acid in the solution during bromination must be a t least 0.48 if the reaction is t o be complete in one minute. I n Table V, i t is evident that the weaker acid concentrations will effect complete bromination if a longer reaction period be allowed. TABLEv--&CID AND TIME A\*
50 cc. water.
"21 -Phenol hTorCc. mality 1 0.16 1 0.16 1 0.16 1 0.16 2 0.32 2 0.32
Yo. of expt. 59 60 61 62 63 64 65 2 66
2
0 32 0.32
Time t o liberate iodine, 1 minute
Per cent sol. Cc.
15.00 15.00 15.00 14.98 14.99 15.00 15.00 15.00
Re-
e x c e s s o f B r action- RI NazB r --period sol. 8 2 0 3 sol. Cc. Calc. Obs. for B r Cc. Cc. 15 00 I 5 28 15.28 15.28 15.27 15.28 15.28 15.30
1.6 3.5 3.5 3.7 3.5 3.5 3.5 3.7
1.0 7.5 15.0 15.0 1.0 2.5 5.0 5.0
14.0 11.1 3.7 3.7 5.7 3.7 3.5 3.4
1.0 2.07 1.0 1.69 1 0 0.53 1 . 0 0.51 1 . 0 0.87 1.0 0.54 1.0 0 . 5 2 1 . 0 0.50
Per cent of phenol found 87.54 92.45 99.83 100.03 97.70 99.84 99.9i 100,23
Thiosulfate. . . . . . . . . . . . 0 . 1 '1' B r . , . . . . . . . . . . . . . . . . . . 0 . 1 .V T e m p . . . . . . . . . . . . . . . . . 22'C.
K I . . . . . . . . . . . . . 20 per cent sol. H C I . . . . . . . . . . . . 1 . 2 sp. gr.
If the acid concentration is 0.16 i V the reaction requires 15 minutes for completion, while the reaction is complete in j minutes if the acid concentration be increased to 0.32 A'. C O M B I N E D E F F E C T O F ACID A K D T E M P E R A T U R E
The experiments recorded in Table VI indicate that the bromination of the phenol is not complete in 0.8 1%' acid in one minute's time if the solution is kept a t z o C. Fifteen minutes' shaking is required a t this low temperature to complete the reaction. I n these low-temperature experiments a precipitation of the tribromphenol did not begin until the solution had been shaken for three minutes. TABLEVI-REACTIOX
P E R I O D WITH
50 cc. water, 5 cc. HC1
KO. of Phenol e x p t . sol. Cc. 67 68 69 70
14.99 15.00 15.00 15.00
STRONG A C I D
A T L O W TEMPERATURES
Time t o liberate iodine, 1 m i n u t e Temperature a t end of titration, 2' C. Per cent ex- R e cess B r sol. action K I c _ _ period sol. Na&Os Br Obs. for B r Cc. Cc. sol. Cc. Cal. 15.28 15.28 15.28 15 28
3.3 3.5 3.5 3.5
12.5 5.3 4.1 3.4
K I . . . . . . . . . . . . . . 2 0 per cent. H C 1 . . . . . . . . . . . . . 1.2 sp. gr. P h e n o l . . . . . . . . . . . 0,09843 N.
1 5 10 15
1 1 1 1
1.84 1.00 0.60 0.50
Per cent phenol found 90.88 98.27 99.41 100.10
Thiosulfate. . . . . . . . 0.1 T i B r . . . . . . . . . . . . . . . . 0.1 S T e m p . . . . . . . . . . . . . 2 2 ' C.
Tables V and VI show that the two principal factors in completing the bromination of the phenol in one minute's time are acid concentration and temperature. The temperature should be roughly 20-30' C. and the acid concentration 0 . 5 LV t o 1.0.-\I EXCESS
BRO M IDE
IN
-
B RO M I D E B RO MAT E
S 0 L U T I 0 Sv
A standard bromide-bromate solution (Koppeschaar's solution) consists of 2.76 grams of potassium 1
TVright. Chem. News, 21, 103.
T H E J O U R N A L OF I N D U S T R I A L A N D E N G I N E E R I N G C H E M I S T R Y
3 92
bromate and 43.1 grams of potassium bromide per liter. According t o the reaction: 5KBr 6HC1 = 6KC1 3Br, 3H,O KBrO, 2.76 grams of potassium bromate require 9.8 grams of potassium bromide, consequently the amounts stated above give an excess of 340 per cent of bromide. Such a large excess of. potassium bromide seemed to the authors unnecessary in this method, although i t has been pointed out by S. J . Lloyd1 t h a t the rate of formation of tribromphenol bromide is decreased b y the addition of potassium bromide. A bromidebromate solution was, therefore, made up containing 2.76 grams of potassium bromate and 15 grams of potassium bromide per liter, giving a n excess of 5 0 per cent of potassium bromide over, that required b y the above equation.
+
+
+
TABLEVII-EXCESS BROMIDEI N 50 cc. water, 5 cc. HCl.
-
THE BROMIDE-BROMATE SOLUTION
Reaction period for Br, 1 minute Time t o liberate iodine, 1 minute
HC1
No.
of expt.
71 72
73 74 75
+
Per cent excess B r sol.
HzO h'or- Phenol Br __c7 Cc. Cc. mality sol. Cc. sol. Cc. Cal. Obs. 15.00 15.20 50 5 0.8 3 3 50 5 0 . 8 15.00 15.20 3 3 14.99 15.19 50 5 0 . 8 3 3 50 5 0.8 15.00 16.47 12 12 15.98 17.74 12.7 12.7 50 5 0.8 Thiosulfate..
Phenol..
Temp.
........ 0,09843 h'
0.1 0.1 0.1 0.1 0.1
0.42 0.44 0.44 1.67 1.99
100.1 99.9 99.9 100.1 100.1
........... 0.1 N .................... 0.1N
K I . . . . . . . . . . . . . 20 per cent sol. HC1.. . . . . . . . . . . 1.2 sp. gr.
Br
Per cent of sol.Na&03 phenol Cc. Cc. found
KI
. . . . . . . . . . . . . . . . 2 2 O C.
The results recorded in Table VI1 show t h a t a decrease of 2 9 0 per cent in the excess potassium bromide does not affect the accuracy of determinations by this method. The tribromphenol precipitate was of the same flocculent nature as t h a t from the bromidebromate solution with the higher potassium bromide content, and in no instance was there a yellowish tinge, indicative of tribromphenol bromide, in the white precipitate. The results in the table show t h a t if any tribromphenol bromide was formed i t was quantitatively reduced by the free hydriodic acid.2 It may be mentioned t h a t if one overshoots with the thiosulfate back titration b y the use of the bromidebromate solution directly, should not be practiced. The free bromine oxidizes the thiosulfate and tetrathionate t o sulfate and errors amounting t o 2 2 per cent of the quantity of bromide-bromate solution used for back titration may be introduced. To avoid the necessity of a standard iodine solution for back titration, we have used the following method and have found i t both accurate and convenient: To I O cc. of water in a test tube, add I cc. of hydrochloric acid, 4 or 5 drops of bromide-bromate solution, carefully measured from the burette, a few drops of 2 0 per cent solution of potassium iodide, and as soon a s the iodine is liberated, wash the liquid from the test tube into the determination bottle. If a sufficient quantity of the bromide-bromate solution has been taken, the blue starch iodide color is a t once restored, and titration with the thiosulfate can be continued t o an accurate end point. 1 Jour.
A m . Chem. Soc., 27, 15 (1905).
* Lloyd, Jour. A m . Chem. Soc , 27,
15 (1905).
Vol. 5 , No. 5
DIRECTIONS
( I ) Solutions required are 0.1 N sodium thiosulfate (24.8 grams per liter); 0.1 N bromide-bromate ( 2 . 7 6 grams KBrO, and 15 grams KBr per liter); 2 0 per cent ,potassium iodide and 0.5 per cent starch solution. The bromide-bromate solution must be compared with the sodium thiosulfate by adding acid and potassium iodide and titrating the iodine set free. (2) Into a 5 0 0 cc. bottle, fitted with a ground glass stopper, p u t 50 cc. water, 5 cc. hydrochloric acid (sp. gr. 1.2) and then add 15 cc. of the unknown phenol solution which is to be determined and which has been diluted t o approximately 0.1 N. If the solution is weaker than 0.1 N, no previous dilution is necessary. Add, while shaking slowly, enough 0.1 N bromidebromate solution to give the solution in the bottle a sIight yellow color which remains permanently. The temperature of the liquid a t this point should be about 2 2 O C. Place the stopper in the bottle, giving it a sharp twist t o bring the surfaces firmly together and shake continuously for one minute. Remove the stopper, add to the solution in the bottle 0.5 cc. potassium iodide solution (20 per cent), replace the stopper and again shake continuously for one minute. Wash down the stopper and sides of the bottle and titrate the solution with 0.1 N sodium thiosulfate, using starch solution as indicator. The starch solution should not be added until enough sodium thiosulfate has been run in t o make the solution almost colorless. It is well, before the blue color of the starch iodide has entirely disappeared, t o replace the stopper and shake the bottle vigorously. Care should be taken not t o overshoot the end point and a standard of tribromphenol in water with excess of sodium thiosulfate should be at hand for comparison. As a further precaution the solution may be left a faint blue as less error is introduced in this way than by titrating to colorless with the attendant danger of overshooting. I n case of overstepping the end point, a solution containing free iodine should be used for titrating back. If a solution containing or yielding free bromine be used for back titrating, concordant results are not obtained. The difference between the quantity of sodium thiosulfate used and the known quantity of bromidebromate solution added, gives the amount of bromidebromate used up in the formation of tribromphenol. Each cubic centimeter of 0.I N bromide-bromate solution is equivalent to 0.0015675 gram phenol. SUMMARY
The results show that phenol may be determined rapidly by bromination to within a n error of o.oooo5 gram. 2. The quantities of reagents for determining phenol as generally recommended have been decreased without sacrificing either speed or accuracy: (a) Two per cent excess of free bromine is sufficient for the complete bromination of phenol in I minute in an acid solution 0.8 N a t 22' C. ( b ) A large excess (over 50 per cent) of potassium iodide above that necessary for complete reduction I.
May, 1913
T H E JOURNAL OF I N D U S T R I A L A N D ENGINEERI-VG CHE-IIIITTRY
of the free bromine according t o their equation is not required, (6) I t has been shown t h a t not more than 50 per cent excess potassium bromide over t h a t called for by the equation at the top of p. 392 need be used in making the standqrd bromide-bromate solution. The U. S. P. recommends a n excess of 340 per cent. 3. Complete liberation of the iodine by the free bromine may be effected in one minute, if thorough diffusion be obtained by sufficient shaking. 4. The acidity of the solution in which the tribromphenol is precipitated must not fall below 0.48 r\i if the bromination is t o be complete in one minute. An acidity of 0.5-1.0 i V is recommended. 5 . Low temperatures have a retarding influence on the rapid formation of tribromphenol: 20-30’ C. is recommended. DEPARTMENT O F INDUSTRI.4L RESEARCH UNIVERSITY OF KANSAS I..4WRENCE
PETROLEUM ANALYTICAL METHODS’ B y S. I?. SADTLER
~
The object of this paper is t o discuss the following: Can the presence of oxygen in petroleum and asphalts be established by a direct method of ultimate analysis? To get the full import of this question, a few words of introduction are needed, bearing upon the subject of what those interested in the chemistry of petroleum and asphalt know with regard t o this matter of the presence of oxygen in substances of these two classes. Hoefer2 gives a list of 59 ultimate analyses of petroleums from all countries. I t is true t h a t more than half of these are the earlier analysesof St. Claire Deville and Boussingault in which only carbon and hydrogen were determined and the balance needed t o make I O O was assumed t o be oxygen, but in a large number of more recent analyses both the sulfur and the nitrogen when present have been directly determined and the balance then ascribed to oxygen. Notably in Russian oils and Japanese oils, both analyzed in recent years and noting the sulfur a n d nitrogen, has this presence of oxygen been recorded. Rakusins also quotes more recent analyses of Russian petroleums b y Charitschkoff and by Nastjukoff, who find from 0.4 t o 2.5 per cent of oxygen and what is of interest, note t h a t the percentage of oxygen increases in the heavy petroleums and residues with the specific gravity. But we are not obliged t o base our belief on the presence of oxygen in petroleums on calculations made from ultimate analyses. The discovery of the petroleum acids b y Hell and Medinger in Roumanian oils and phenols and of the naphthene-carboxylic acids by Markownikoff and Oglobin has given us a n explanation of the presence of oxygen and justified the assumptions made from the ultimate analyses. With the natural asphalts, the case is different from t h a t of petroleums. Although earlier ultimate analyses of asphalts gave large percentages of oxygen, Paper presented a t the Eighth International Congress of Applied Chemistry, before the Section on Fuels and Asphalt, Sept. 6, 1912. 2 Das Erdoel und seine v e r z ~ a ~ d t e 2te n , Auf., Seiten 55 und 56. 3 Die Untersuchung d e s Erdoels una! seine Producte, 1906, 7 7 . 1
393
i t was because the presence of sulfur in them had not been recognized and the oxygen was supposed, with the carbon and hydrogen, t o make up the ash-free bitumen. However, Kohlerr gives several analyses of natural asphalts by Day and Bryant and by Kayser, in which a small percentage of oxygen is given a s present along with a larger percentage of sulfur. Both Clifford Richardson and Prof. S. F. Peckham, eminent American authorities on asphalt, have taken the position t h a t not only is sulfur a distinctive element for natural asphalts, but equally that oxygen is t o be considered as foreign t o natural asphalts. Besides the natural asphalts, we have also t o note the artificial asphalts, obtained from ‘ petroleum, either by simple removal of the volatile portions o r by some form of treatment with oxygen or sulfur at high temperatures. To the first class belong such products as “ D grade asphalt,”2 made from California asphaltic petroleum, and “ Baku Pitah”3 and to the second class Ventura Flux, Byerlite and Sarco asphalt. Of these last mentioned products, Byerlite and Sarco asphalt have been made irom liquid petroleum-residuums b y the action of a current of air, either drawn through or forced through a t temperatures ranging from 380‘ F. (193.3’ C.) to 500’ F. (287.70C.). The action of the heated air may have two different effects4 according t o temperature and rapidity or quantity of air passed through. The oxygen may cause splitting off of hydrogen in the form of water, with condensation of the hydrocarbons affected, or the oxygen may be fixed, forming products of oxidation which remain, in either case resulting in thick semi-solid or solid products. Not only would it be very desirable from a scientific point of view t o determine which of these reactions has taken place, or whether both have united in the formation of the solid asphalt-like products obtained, but the matter has been the subject of investigation in connection with patent litigation over rival processes. Of course, direct determinations of carbon, hydrogen, sulfur and nitrogen may and (do leave varying deficiencies t o be charged up t o oxygen, but it would be desirable to be able t o confirm these calculations by a direct determination of the oxygen in the product. No such method has thus far come into common use. The method of Baumhauer, either in its earlier form or in its later form, using a weighed quantity of dry silver iodate and requiring first a current of hydrogen, then of nitrogen and finally of hydrogen again, has not been favorably commented Ion by those who have tried it. The method of Mitscherlich of burning with mercuric oxide is also intended. t o give the oxygen a t the same time t h a t the carbon and hydrogen are obtained, but this method does not seem t o have worked satisfactorily in the hands of those who have referred to i t and has not been adopted by chemists. The process which I desire t o present t o those in1 Chemie u n d Technologie der .\’atiirlichen 1904, 81. 3 4
uttd Kiirtstlichen Asfihalte,
Clifford Richardson, “The Modern Asphalt Pavement,” [2] 1908, 263. Ibid., 271. Hofer, p. 85.
